similar to: How safe is mbox_very_dirty_syncs?

I've got some of our busiest users now using Dovecot instead of UW-IMAP with mbox-format mailboxes and predominantly Outlook Express as the client. Doing some perfomance monitoring (with Solaris 8 process accounting), it looks like Dovecot needing to read only 30-50% of the disk blocks needed by UW-IMAP, but I was hoping for better! What seems to happen, is that most of the connections

I've been struggling with an odd bug in the latest 1.0-stable (20050427) release. Sometimes Dovecot seems to forget that a message has been read or deleted and marks it unread (or not deleted). As far as I've been able to determine, it seems to happen when 1) There have been several recent deliveries 2) Possibly some of those deliveries are of multiple messages 3) Possibly they are read

I've tried this on both Solaris 8 and SuSE Enterprise 9 (64-bit). I get a assert-crash when using a gzipped mbox folder . OK Logged in. . SELECT test.gz * FLAGS (\Answered \Flagged \Deleted \Seen \Draft $Label4 $Label2 $Label1 $Label3 $Label5) * OK [PERMANENTFLAGS ()] Read-only mailbox. * 167 EXISTS * 0 RECENT * OK [UNSEEN 2] First unseen. * OK [UIDVALIDITY 1096038620] UIDs valid * OK

Hmm, this is peculiar! This mailbox has been unchanged on my test server since September. I've been using it to test each new version of Dovecot just after I've compiled it. When beta1 was released, I opened this mbox with no problems. However, when I upgraded to beta2, I got "file size unexpectedly shrinked in mbox file /export/mail/a/24/vis99003/INBOX (16895967 vs 16899267)".

I've been seeing lots of index cache file errors (using mbox on Solaris 8 sparc 64-bit, but with 32-bit Dovecot) since I switched my account to Dovecot 1.2.0 (we're still on 1.0.15 mostly, but I'm hoping to upgrade to 1.1.17+ this summer, or 1.2.x if it's stable enough). e.g. Error: Corrupted index cache file <path>/.imap/INBOX/dovecot.index.cache: field index too large (47

It seems that Dovecot gets confused when presented with a header like: From: someone at somewhere.org i.e. with leading whitespace and no "friendly name" This shows up as ""@MISSING_DOMAIN in clients, such as Pine, that believe what Dovecot tells them rather than parsing the headers themselves (e.g. Thunderbird). Here's a sample IMAP session: xyz FETCH 728 ENVELOPE

Is it possible to have folders on read-only partitions in Dovecot 1.0-stable? We have a NetApp fileserver which would probably be accessed via NFS (though iSCSI is possible) and which stores readonly "snapshots" of every directory accessible as .snapshot (but usually hidden from the operating system). It would be very nice to set up a "snapshot" namespace something like

Timo, I've noticed a couple of extra compilation errors in last night's snapshot compared to the snapshot I downloaded on Tuesday, probably due to the int -> bool change :- (gcc 3.3.2 on Solaris 8) mbox-sync.c: In function `mbox_sync': mbox-sync.c:1568: warning: comparison is always false due to limited range of data type index-search.c: In function `search_index_arg':

I've got a problem with subscriptions in 1.0-stable when migrating from UW-IMAP. I've modified the SUBSCRIPTION_FILE_NAME entries as suggested in the Wiki http://wiki.dovecot.org/Migration and included the "~/mail/" etc. hidden namespaces. The .mailboxlist files contain entries like "~/mail/somefolder" and the clients have "~/mail/" set as a prefix. In

We upgraded from Dovecot 1.0.15 to 1.2.5 last night, on Solaris 10 using mboxes, mostly without issues. However I had to trash the index/cache files (too many folders were showing corruption issues which is especially bad for Prayer Webmail ".prayer" folders that store preferences; Prayer sees a disconnection as the folder being missing!). I've had one imap process panic in mailbox

I've had complaints from users about "MISSING_DOMAIN" in header fields and I've tracked it down to having "." in an unquoted display name. By my reading of RFC2822 (especially section 4.1) A N Other <a.n.other at somewhere.org> "A. N. Other" <a.n.other at somewhere.org> are legal, but A. N. Other <a.n.other at somewhere.org> is

I've noticed that my dovecot.index.cache file has got very large (24MB, which is more than 1/3 of my INBOX size). I've tried deleting (well, renaming) the dovecot.index.* files, and re-opening my INBOX, and the new dovecot.index.cache is only 90KB so far. I'm using dovecot-1.0-stable and mbox format. My mail_cache_fields and mail_never_cache_fields are set to the default. Looking

I've got several users testing Dovecot IMAP at the moment, but some of them are also accessing their mail via UW-IMAP at other times. (I've warned them about accessing the same folder simultaneously via UW-IMAP and Dovecot!) In addition, we have special folders "AutoSpam" and "Trash" that pruned using Washington's "mailutil" automatically every so

Hi all, We moved our Dovecot installation first of all to have all folders and inboxes on NetApps via NFS (with indexes local) then from a physical Solaris 8 4-way UltraSparc server running Dovecot 1.2.5 (32-bit) to a virtualised Ubuntu 8.04 64-bit server with a single virtual processor (in VMWare ESX 3.5) running Dovecot 1.2.8 (64-bit). Since the change, many of our users have experienced

I've been having problems with Pine 4.62 talking to Dovecot 1.0 stable. It's fine when the folder collection path is empty, but if I have "~/mail/" or "mail/" it gives access errors (I'm using the "~/mail/", "mail/" etc. hidden namespaces for backwards compatability). It seems to me that Dovecot is not replying correctly to a "LIST"

We've had another random imap process crash. This is with the original 1.2.5 imap (I haven't applied the patch for two processes creating an index simultaneously): > Oct 03 13:24:56 imap-login: Info: Login: user=<xxxxxxxx>, method=PLAIN, rip=134.225.1.46, lip=134.225.16.6 > Oct 03 13:25:59 IMAP 6067 xxxxxxxx 134.225.1.46 : Info: delete: uid=483, msgid=<xxxxxxxx> > Oct

Anybody else seen this one? We're running Dovecot 1.0.rc7 on a different port on our live server (we're generally live with 1.0.beta5) and after a HUP to reload the user database (in a passwd-file) this morning we got:- > dovecot: Sep 22 07:45:05 Info: IMAP 14482 someuser xxx.xxx.xxx.xxx : Disconnected in IDLE > dovecot: Sep 22 07:45:05 Warning: imap-login: SSL_read() syscall

Has anybody managed to have Dovecot authenticate to Microsoft Active Directory? I've tried to get it to work using LDAP and the following settings dovecot.conf: userdb = ldap /etc/dovecot-ldap.conf passdb = ldap /etc/dovecot-ldap.conf dovecot-ldap.conf: hosts = rdg-home.ad.rdg.ac.uk dn = cn=bind-user,cn=users,dc=rdg-home,dc=ad,dc=rdg,dc=ac,dc=uk dnpass = XXXXX ldap_version = 2 base =

I think I've found a minor bug in the LSUB command: . LSUB "" "*" gives ... * LIST (\Noselect) "/" "#maildir" ^^^^ * LSUB () "/" "#maildir/test4" where I have a namespace with prefix "#maildir/". Here's a patch to make it not list the namespace prefix in an LSUB command (or should it list it as 'LSUB

Hi Timo, Just a quick question: do index files still need removing when upgrading from 1.0-stable (as in the testXX series) or will Dovecot regenerate them automatically? I'm expecting to go with 1.0-stable on Tuesday for my 20,000 users (as that's what I've been testing), but if we see lots of index issues, I may want to upgrade to alpha1 (or later) fairly soon afterwards and it